JP2013241446A - N-terminal derivatization of protein with polysaccharide - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method for producing an N-terminal derivative of a protein in which a polysaccharide, preferably having at least terminal sialic acid units and preferably essentially comprising only sialic acid units, is reacted at an N-terminus of a protein or a peptide under controlled conditions to produce the N-terminal derivative.SOLUTION: A method for producing a purified N-terminal derivative of a protein or a peptide comprises: (i) reacting a polysaccharide with an amine group of an N-terminus of the protein or the peptide in aqueous acidic solution to produce an N-terminal derivative; and (ii) purifying the resultant N-terminal derivative in an aqueous solution of a higher pH than in step (i).

Description

  The present invention preferably comprises a polysaccharide having at least a terminal sialic acid unit, and preferably consisting essentially of sialic acid units, reacting at the N-terminus of a protein or peptide under controlled conditions to produce an N-terminal derivative. The present invention relates to a method for producing an N-terminal derivative of a protein. Derivatives are useful for improving the pharmacokinetics and pharmacodynamics of proteins and peptides.

  Polysialic acid (PSA) is a naturally occurring unbranched polymer produced by certain bacterial strains and in certain mammalian cells. Polysialic acid can be produced at various degrees of polymerization from n = about 80 or more down to n = 2 by limited acid hydrolysis, or digestion with neuraminidase, or fractionation of the natural bacterially derived form of the polymer. In recent years, the biological properties of polysialic acid, particularly α-2,8 linked homopolymeric polysialic acid, have been utilized to alter the pharmacokinetic properties of proteins and low molecular weight drug molecules. Polysialic acid derivatization dramatically improves blood half-life for many therapeutic proteins, including catalase and asparaginase, resulting in undesirable (and sometimes unavoidable) pre-exposure to therapeutic proteins Such a protein can be used even in the case of facing (Non-Patent Document 1, Non-Patent Document 2). Alpha-2,8 linked polysialic acid provides an attractive alternative to PEG, an immunological that is a natural part of the human body and degrades to sialic acid, a non-toxic saccharide via tissue neuraminidase Invisible biodegradable polymer.

The present inventors have so far described a method of linking a polysaccharide and a therapeutic agent such as a protein (Patent Document 1, Patent Document 2). Some of these methods are by chemical derivatization of the “non-reducing” end of the polymer to create a protein-reactive aldehyde moiety that reacts with primary amine groups. Non-reducing sialic acid terminal units contain vicinal diols and are therefore easily (and selectively) oxidized by periodate, highly reactive to proteins, and reductive amination and other Monoaldehyde forms can be obtained that contain reaction factors suitable for linking proteins through chemical reactions. This reaction is shown in FIG.
A) shows the oxidation of colominic acid (α-2,8 linked polysialic acid from E. coli) by sodium periodate, forming a protein reactive aldehyde at the non-reducing end,
B) Selective reduction of the Schiff base with sodium cyanoborohydride to form a stable irreversible covalent bond with the protein amino group.

  Due to the multifunctional nature of the protein, PSA and PEG conjugation strategies irreversibly result in a mixture of chemically different molecular entities. For example, the reaction between colominic acid and the side chain of an amino acid may produce an unintended by-product during the conventional binding reaction described above. These may be sufficient to make it difficult to produce chemically defined complexes required by regulatory authorities for therapeutic use in humans and animals.

Site-directed approach of protein and poly (ethylene glycol) attachment that allows the preparation of essentially homogeneous PEG-protein derivatives with a single PEG chain attached to the amine terminus of the protein Is described in Non-Patent Document 3. This selectivity is achieved by performing reductive alkylation of the protein with PEG-aldehyde at a lower pH than usual. However, this method was applied to conjugation with PEG only.

  The chemical role of PEGylation cannot be applied to polysialation due to differences in the physiochemical properties of these molecules. PSA is an acid labile polymer and is stable for several weeks near neutral pH (Figure 2.2). The results in FIG. 2.2 show that at pH 6.0 and 7.4, CA is stable for 8 days, and at pH 5.0, degradation is slow (92% of the initial MW after 48 hours), pH 4. At 0, degradation is slow (70% of initial MW after 48 hours). While polysialic acid is highly hydrophilic, PEG is an amphiphilic molecule in nature. When polysialylation is performed using the conditions used for PEGylation, protein aggregation and precipitation are often observed.

US Pat. No. 5,846,951 (A) International Publication No. 0187922 (A)

Fernandes and Gregoriadis, 2006 Jain et al., 2003, 2004 Kinstler et. Al., 2002

  One of the biggest challenges remaining in the development of protein formulations is to address the physical and chemical instabilities of proteins. Protein formulations usually must be stored under cold conditions or lyophilized to achieve an acceptable shelf life. Excipients (eg, stable buffers, sugars, polyols, surfactants, salts, PEGs, polymers, metal ions and amino acids) to improve protein stability and give protein formulations greater stability and shelf life And structural changes are utilized. There remains a need to improve the formulation of protein-PSA complexes.

  In accordance with a first aspect of the present invention, the inventors have provided a method for producing a purified N-terminal derivative of a protein or peptide comprising: (i) a polyhydration with an amine group at the N-terminus of the protein or peptide in an acidic aqueous solution. A method for producing a purified N-terminal derivative of a protein or peptide by reacting a saccharide to produce an N-terminal derivative, and (ii) purifying the obtained N-terminal derivative in an aqueous solution having a higher pH than in step (i) I will provide a.

  In the present invention, the inventors have made it possible to prepare a stoichiometrically defined site-specific complex of a polysaccharide, preferably polysialic acid (PSA), with the protein N-terminus. And the results of applying a scalable chemical approach. Typically this is due to reductive alkylation with PSA-aldehyde in an aqueous environment. The N-terminal linkage is advantageous because it is believed to minimize any undesirable interference between the polysaccharide and the remaining amino acid residues in the protein.

  The method requires that the pH be returned to neutral pH immediately after the polysaccharide is fully degraded by exposure to a derivatization or otherwise derivatization buffer in an acidic environment. The method also requires pH control and optionally the presence of formulation additives during the reaction to prevent protein precipitation.

  The present invention enables the synthesis of defined chemically homogeneous and effective conjugates of proteins that are highly useful in the pharmaceutical industry.

  In the present specification, the terms “protein” and “peptide” are used interchangeably.

  Proteins react primarily with the N-terminal amine group. This reaction is performed by the pH conditions of the reaction that selectively promotes chemical reactions at the N-terminus of the protein. Figure 2.1 shows protein derivatization with PSA. Shown is the selective reduction of the Schiff base with cyanoborohydride to form a stable irreversible covalent bond with the N-terminal amino group.

  An acidic aqueous solution is a reaction medium in which a derivatization reaction is performed. This can be a buffer solution, such as sodium acetate. The reaction medium is aqueous.

  In the method according to the first aspect of the present invention, the purification step (ii) needs to be performed at a higher pH than the step (i) which is the derivatization step. Preferably, the purification step is performed at a substantially neutral pH. When purification is performed at the same (acidic) pH as in the derivatization step, the polysaccharide is considered to be susceptible to degradation. By acidic pH we intend a pH of less than 7. Preferably, in step (i) of the method, the pH of the acidic aqueous solution is in the range of 3.0 to 6.5, more preferably in the range of 4.0 to 6.0. The pH of the solution used in step (ii) of the method is less acidic than the pH of the solution in the first step. This pH is preferably substantially neutral and has a pH in the range of 6.5 to 9.5, preferably 6.5 to 8.5, and most preferably 6.5 to 8.0.

In one embodiment of the invention, the polysaccharide is activated prior to reaction with the protein or peptide to produce an activated derivative. Typically, the activated derivative of the polysaccharide has a reactive aldehyde group and step (i) is carried out under reducing conditions. Boron hydride may be used to obtain reducing conditions. The N-terminus of a protein reacts with a reactive aldehyde group and produces an adduct that upon reduction produces an N-terminal derivative of the protein or peptide.

  Reactive aldehydes can be produced by selective oxidation of polysaccharides using periodate.

Preferably, the activation of the polysaccharide should be carried out under conditions such that there is substantially no medium chain cleavage of the backbone of the long chain (polymerization) starting material, i.e. there is substantially no decrease in molecular weight. The activation step is typically conditional on the aldehyde moiety being at the end of the polysaccharide. An enzyme capable of performing this step may be used. Most conveniently, the oxidation is a chemical oxidation. The reaction is polymer-based perruthenate
You may carry out by immobilization reagents, such as. The most direct method is performed using a lysis reagent. The oxidizing agent is suitably perruthenic acid, or preferably periodate. Oxidation may be carried out at a periodate ranging from 1 mM to 1 M, a pH ranging from 3 to 10, a temperature ranging from 0 ° C. to 60 ° C., and a time ranging from 1 minute to 48 hours.

Reduction conditions suitable for the derivatization reaction may utilize hydrogen with a catalyst, preferably a hydride such as borohydride. These may be fixed with Amberlite (trademark) supported borohydride or the like. Preferably, such as sodium borohydride at a concentration in the range of 1 μM to 0.1 M, a pH in the range of 5.0 to 10, a temperature in the range of 0 ° C. to 60 ° C., and a period in the range of 1 minute to 48 hours. Alkali metal hydrides are used as reducing agents. Reaction conditions are selected so that pendant carboxyl groups on the starting material are not reduced. Other suitable reducing agents are cyanoborohydrides under acidic conditions such as polymer supported cyanoborohydrides or alkali metal cyanoborohydrides, L-ascorbic acid, sodium metabisulfite, L-selectride, triacetoxy hydrogenation Such as boron.

  Other activated derivatives of polysaccharides are useful in the present invention, including those with pendant functional groups such as NHS, as described in our previous patent application, International Patent Application No. WO06 / 00540. May have sex.

  Preferably the polysaccharide is an anionic polysaccharide such as polysialic acid (PSA), heparin or chondroitin sulfate. Most preferably, the polysaccharide is PSA. However, in the present invention, this preferred polysaccharide starting material may contain units other than sialic acid in the molecule. For example, the sialic acid unit may replace other saccharide units. Preferably, however, the polysaccharide consists essentially of sialic acid units. Preferably they are linked 2 → 8 and / or 2 → 9.

  Preferably, the polysaccharide starting material has at least 2, more preferably at least 5, more preferably at least 10, for example at least 50 saccharide units. For example, the polysaccharide may comprise at least 5 sialic acid units.

The polysialic acid is derived from any source, preferably a natural source such as a bacterial source (eg E. coli K1 or K92, group B meningococci, or even milk or N-CAM). The sialic acid polymer can be a heteropolymer such as Neisseria-N. Meningitidis group 135 or group V.

  The polysialic acid can be in the form of a salt or free acid. The polysialic acid can be in a hydrolyzed form that reduces the molecular weight after recovery from a bacterial source.

  The polysaccharide, preferably polysialic acid, can be a wide range of molecular weight materials such as having a polydispersity of greater than 1.3, such as 2 or more. Preferably, the polydispersity of the molecular weight is less than 1.3 or 1.2, more preferably less than 1.1, such as 1.01.

  Various methods known in the art may be used to purify the N-terminal derivative in step (ii) of the method of the first aspect of the invention. Examples of suitable purification methods include HIC (hydrophobic interaction chromatography), SEC (size exclusion chromatography), HPLC (high performance liquid chromatography) and AEX (anion exchange chromatography).

  In the process of the present invention, the reactive aldehyde is preferably at the non-reducing end of the polysaccharide. However, the reactive aldehyde may be provided at the reducing end of the polysaccharide. Suitable chemistries for preparing polysaccharides with reactive aldehydes at the reducing end of the polysaccharide are described in our previous patent application, International Patent Application No. WO05 / 016974. This process involves a pre-selective oxidation step followed by a reduction step followed by an oxidation step to produce a compound having an aldehyde at the reducing end and the passivated non-reducing end.

  The present invention is particularly useful for the production of derivatives of therapeutic proteins. The protein can be, for example, obestatin, leptin, interferon, FSH, galactosidase or DNase.

The formulation additive may be present in the aqueous solution in either or both of step (i) and step (ii) of the method according to the first aspect of the invention.

By formulation additive we intend an excipient capable of stabilizing proteins or peptides either internally or externally as described in Wang et al. (1999). Excipients can be stabilizers, solubilizers or metal ions. Suitable examples of formulation additives include one or more of buffers, stabilizers, surfactants, salts, polymers, metal ions, sugars, polyols, or amino acids. These may be used alone or in combination.

  Stabilizers typically act by destabilizing a denatured protein that results in an increase in Gibbs free energy change for protein unfolding. Stabilizers are preferably sugars or polyols such as sucrose, sorbitol, trehalose, glycerol, mannitol, lactose and ethylene glycol. The stabilization buffer is sodium phosphate.

It is preferred that the solubilizer is a surfactant, preferably a nonionic surfactant. Preferable examples include Tween 80, Tween 20, Tween 40, Fluoronic.
F68, Brij35 and Triton X100.

The metal ion is preferably divalent. Suitable metal ions include Zn ²⁺ , Ni ²⁺ , Co ²⁺ , Sr ²⁺ , Cu ²⁺ , Ca ²⁺ , Mg ²⁺ and Fe ²⁺ .

  The formulation additive may also be a polymer selected from PSA, PEG or hydroxy-β-cyclodextrin.

  Suitable amino acids and amino acid derivatives for use as formulation additives include histidine, glycine, other similar amino acids and sodium aspartate.

  The polysialic acid population having a broad molecular weight distribution may be fractionated into fractions having a lower polydispersity, that is, fractions having different average molecular weights. Fractionation is preferably performed by anion exchange chromatography using a basic buffer suitable for elution as described in our previous international patent applications WO2005 / 016794 and WO2005 / 03149. . The fractionation method is suitable for polysialic acid starting materials and derivatives. This technique may therefore be applied before or after the essential process steps of the present invention. Preferably, the polydispersity of the resulting polysialic acid derivative is less than 1.1.

  According to a second aspect of the invention, the inventors comprise a composition comprising a population of polysialic acid derivatives of a protein, wherein the derivative comprises 2 to 200 sialic acid units, the population substantially comprising Compositions comprising a polysialic acid derivative population of proteins consisting only of N-terminal derivatives of proteins are provided.

  By population we mean that there are two or more polysialic acid derivatives in the composition. The derivatives may contain the same or different numbers of sialic acid units. Preferably, the polydispersity of the polysialic acid in the composition is less than 1.3, more preferably less than 1.1.

  In the population, virtually all proteins are derivatized with only the N-terminal amine. This contemplates that at least 85%, preferably at least 90%, most preferably at least 95% of the proteins in the population are derivatized by PSA with only the N-terminal amine. Techniques known in the art such as peptide mapping and Edman degradation can be used to determine the degree of derivatization at the N-terminus.

  The protein can be any protein with therapeutic use, such as obestatin, leptin, interferon, FSH, galactosidase or DNase.

  When the protein is FSH, the derivative typically contains 75 to 200 sialic acid units.

  When the protein is α-galactosidase, the derivative typically contains 20 to 150 sialic acid units.

  When the protein is a DNase, the derivative typically contains 2 to 120 sialic acid units.

  When the protein is IFN, the derivative typically contains 80 to 180 sialic acid units. Preferred polysialic acids are those detailed above in other embodiments of the invention.

The polysialic acid may be coupled to the protein directly, ie as shown in FIG. 1, or via a linker. Suitable linkers are derived from reagents containing N-maleimide, vinyl sulfone, N-iodoacetamide, orthopyridyl or N-hydroxysuccinimide functional groups. The linker may be biologically stable or biodegradable and may comprise, for example, a polypeptide or a synthetic oligomer. The linker may be derived from a bifunctional group-containing reagent as further described in International Patent Application No. WO2005 / 016973. For example, a suitable bifunctional reagent is Bis-NHS. This reagent has the general formula Z—R ¹ —Z, where each Z is a functional group, which may be the same or different and R ¹ is a difunctional organic group. obtain. Preferably, R ¹ is alkanediyl, arylene, alkarylene, heteroarylene and alkylheteroarylene, any of which is substituted with a carbonyl bond, ester bond, sulfide bond, ether bond, amide bond and / or amine bond and / or Or may be interposed). C ₃ -C ₆ alkanediyl is particularly preferred. Most preferably, R ¹ corresponds to a suitable part of a suitable bifunctional reagent.

  The polysialic acid derivative has the general formula (I)

Wherein m is at least 1,
HNB is B-NH ₂ from the N-terminal amine of the protein,
L is a chemical bond or linking group, or comprises a polypeptide or a synthetic oligomer;
GlyO is a sialic acid unit,
The linking group (if present) has the general formula —Y—C (O) —R ¹ —C (O) —, where Y is NR ² or NR ² —NR ² , and R ¹ is as described above. A difunctional organic group, and R ² is H or C _1-6 alkyl).

  In this aspect of the invention, the protein binds to the non-reducing end of the polysaccharide.

When the protein is attached directly to the polysaccharide, the L group is a bond. However, the L group may alternatively be derived from a reagent containing N-maleimide, vinyl sulfone, N-iodoacetamide, orthopyridyl or N-hydroxysuccinimide. The reagent may have the general formula Z—R ¹ —Z as described above. In this embodiment, L is typically a group

It is.

  Compositions comprising polysialic acid derivatives and diluents, pharmaceutical compositions comprising novel biologically active compounds, and pharmaceutically acceptable excipients also form part of the present invention. The pharmaceutical composition may be administered orally, intravenously, intraperitoneally, intramuscularly, subcutaneously, intranasally, intradermally, topically or intratracheally for human or veterinary use.

  The composition may further comprise formulation additives as detailed above.

  The last aspect of the invention is an N-terminal derivative of a protein or peptide obtainable by the method according to the first aspect of the invention. Also any of the preferred properties of the above derivatives are applicable to this aspect of the invention.

Derivatization of proteins and peptides, etc. provides increased half-life, improved stability, reduced immunogenicity, and / or controlled solubility, and thus bioavailability and pharmacokinetic properties, or actives ), Or the viscosity of a solution containing a derivatized active agent. The new method has particular value for the production of monopolysialylated protein complexes. This is based on the understanding that the lower the pH, the more reactive the N-terminal amine group, the higher the reactivity.

  The invention is illustrated by Examples 1 to 10 and with reference to the accompanying drawings.

1 is a reaction scheme showing prior art activation of non-reducing sialic acid end units. FIG. 2 is a reaction scheme showing prior art reductive amination of the aldehyde moiety of the product of reaction scheme 1a with a protein-amine moiety. FIG. 4 is a reaction scheme showing N-terminal derivatization of proteins. FIG. 6 shows degradation of colominic acid (CA) at different pH using triple detection GPC (Viscotek: RI + RALS + Viscosiometer). It is a figure which shows the characteristic of the GCSF-CA composite_body | complex by SE-HPLC and SDS-PAGE. It is a figure which shows SDS-PAGE and SE-HPLC of a 50 kDa CAO-leptin complex. It is a figure which shows the in vivo clearance of an EPO formulation (intravenous). FIG. 5 shows further in vivo clearance of EPO formulation (subcutaneous). It is a figure which shows the characteristic of the EPO-CA complex by SE-HPLC and SDS-PAGE. FIG. 4 shows in vivo clearance of non-glycosylated EPO relative to polysialylated non-glycosylated EPO (n = 3-4 ±, subcutaneous) FIG. 4 further shows in vivo clearance of non-glycosylated EPO relative to polysialylated EPO (intravenous injection, n = 3-4 ± SEM). FIG. 6 shows characterization of NGEPO-CA complex by SE-HPLC and SDS-PAGE. FIG. 6 shows the characterization of CA-insulin complex by SDS-PAGE (Tris-Glycine (4% -20%) gel). FIG. 2 shows the characterization of purified CAO-insulin complex by SE-HPLC. It is a figure which shows the in vivo effectiveness of the CAO-insulin preparation (0.3U, n = 4 subcutaneously) (about 25g) with respect to an outbred female mouse. FIG. 2 shows SE-HPLC of polysialylated interferon α2b (39 kDa reaction mixture with CA after 24 hours). FIG. 3 shows in vivo clearance of polysialylated interferon (intravenous injection, n = 4 ± SEM). It is a figure which shows SDS-PAGE of complex obestatin. It is a figure which shows SDS-PAGE of the composite_body | complex in Example 10.4. It is a figure which shows SE-HPLC of the composite_body | complex in Example 10.5. It is a graph which shows DNase I activity versus calcium chloride concentration.

Materials Ammonium carbonate, ethylene glycol, polyethylene glycol (8 kDa), sodium cyanoborohydride (over 98% pure), sodium metaperiodate and molecular weight markers were obtained from Sigma Chemical Laboratory (UK). Colominic acid used, chain α
-(2 → 8) conjugated E. coli K1 polysialic acid (average 22.7 kDa high polydispersity 1.34, 39 kDa polydispersity 1.4, 11 kDa polydispersity 1.27) from Camida (Ireland) (Na ¹²⁵ I) was purchased from Amersham (UK). Other materials include 2,4 dinitrophenylhydrazine (Aldrich Chemical Company, UK), dialysis tubing (3.5 kDa and 10 kDa cutoff limits, Medicell International Limited, UK), Sepharose SP HiTrap, PD-10 columns (Pharmacia, UK), Tris-Glycine polyacrylamide gel (4-20% and 16%), Tris-Glycine sodium dodecyl sulfate running buffer and loading buffer (Novex, UK). Deionized water was obtained from an Elgastat Option 4 water purification unit (Elga Limited, UK). All reagents used were analytical grade. A plate reader (Dynex Technologies, UK) was used for spectroscopic measurements in protein or CA analysis. B6D2F1 mice (7-8 weeks old, 20 g body weight) were purchased from Harlan (UK) and acclimated for at least one week before use.

Ammonium sulfate, GCSF (Sll, India), sorbitol, Tween 20, Q
FF (column 1 ml or 5 ml, Amersham Biosciences, UK), sodium chloride, sodium phosphate, Hitrap Butyl HP column (1 ml or 5 ml, Amersham Biosciences, UK), mouse leptin recombinant (Biomyx), erythropoietin (EPO) and non-glycosyl EPO (NGEPO) (molecular weight 30600, SIIL, India), Tris (Sigma, UK), sodium acetate (BDH, UK), sodium phosphate (BDH, UK),
Insulin (Sigma, UK).
1. Protein and colominic acid measurements The quantitative evaluation of polysialic acid (as sialic acid) using resorcinol reagent was performed by the resorcinol method (Svennerholm, 1957) as described elsewhere (Gregoriadis et.al., 1993; Fernandes and Gregoriadis, 1996, 1997). Protein was measured by the BCA colorimetric method or UV absorbance at 280 nm.
2.1 Activation of colominic acid A freshly prepared 0.02 M sodium metaperiodate (NaIO ₄ ) solution (8-fold molar excess) is mixed with CA at 20 ° C. and the reaction mixture is allowed to stand in the dark for 15 minutes. Magnetically stirred. Subsequently, twice the volume of ethylene glycol was added to the reaction mixture to consume excess NaIO ₄ and the mixture was stirred at 20 ° C. for an additional 30 minutes. Oxidized colominic acid was dialyzed sufficiently (24 hours) against 0.01% ammonium carbonate buffer (pH 7.4) at 4 ° C. (dialysis tube with a molecular weight cut-off of 3.5 kDa). Ultrafiltration (molecular weight cut-off greater than 3.5 kDa) was used to concentrate the CAO solution from the dialysis tube. After concentrating to the required volume, the filtrate was lyophilized and stored at −40 ° C. until later use. Alternatively, CA was recovered from the reaction mixture by precipitation (twice) with ethanol.
2.2 Measurement of the oxidation state of CA and derivatives Quantitative evaluation of the degree of oxidation of colominic acid is based on 2,4 dinitrophenylhydrazine (2 which produces 2,4 dinitrophenyl-hydrazone which is slightly less soluble in the interaction with carbonyl compounds. , 4-DNPH). Non-oxidized colominic acid (CA) / oxidized colominic acid (CAO) is added to 2,4-DNPH reagent (1.0 ml), the solution is shaken and then placed at 37 ° C. until a crystalline precipitate is observed. (Shriner et al., 1980). The degree of oxidation of CA (quantitative) is determined by a method based on the reduction of ferricyanide ions to ferric ferrocyanide (Persian blue) (measured at 630 nm) in an alkaline solution (Park and Johnson, 1949). And measured. In this example, glucose was used as a standard.
2.3 Gel Permeation Chromatography Colominic acid samples (CA and CAO) were dissolved in NaNO ₃ (0.2M), CH ₃ CN (10%, 5 mg / ml) and the refractive index on a GMPW _XL column more than twice. (GPC system: VE1121 GPC solvent pump, VE3580 RI detector and verification with Trisec3 software (Viscotek Europe Ltd.)), chromatography was performed. A sample (5 mg / ml) was filtered through a 0.45 μm nylon membrane and run at 0.7 cm / min using 0.2 M NaNO ₃ and CH ₃ CN (10%) as the mobile phase.
Results Colominic acid (CA) (polysialic acid) is a chain-like α-2,8 linked homopolymer of N-acetylneuraminic acid (Neu5Ac) residues (FIG. 1a). Oxidative exposure to colominic acid was performed using 20 mM periodate at room temperature for 15 minutes (Lifely et. Al., 1981). The integrity of the internal α-2,8 linked Neu5Ac residue after periodate treatment was analyzed by gel permeation chromatography and chromatograph obtained for oxidized colominic acid (CAO) and the material was analyzed by native CA. Compared with the ones. Oxidized and native CA showed almost the same elution profile, and it was found that there is no evidence of significant fragmentation of the polymer chain with successive oxidation steps.

  Quantitative measurement of the oxidation state of CA was performed by reducing ferricyanide ions to ferrocyanide (Prussian blue) in alkaline solution using glucose as a standard (Park and Johnson, 1949). Table 1 includes the combined reducing power of oxidized colominic acid in excess of stoichiometry (greater than 100%), ie reducing terminal hemiketal and introduced aldehyde (at the other end). , Which was found to have an apparent aldehyde content of 112 mol%.

Table 1: Intermediates the degree of oxidation of various colominic acids in the double oxidation scheme using glucose as a standard (100%, 1 mole aldehyde per mole glucose, n = 3 ± standard deviation).
3. Preparation of N-terminal protein-CA complex using formulation additives 3.1 Preparation of GCSF-CA complex G-CSF (18.8 kDa) is given as a solution (5% sorbitol, 0.025 mg / ml polysorbate) 1.05 mg / ml in 10 mM sodium acetate buffer (pH 4.0) containing 80) and stored at 2-8 ° C. The required amount of GCSF was placed in an Eppendorf tube and placed on ice. The amount of CA (eg, oxidized or non-oxidized CA, greater than 11 molar excess) added for conjugation is expressed by
CA weight = {protein amount (g) / (protein molecular weight)} × (CA molecular weight) × (molar excess of CA)
Calculated based on

The required amount of CA was weighed. After solubilizing the CA in 10 mM NaOAc, 5% sorbitol (pH 5.5) (where 20% by volume of the final reaction volume was used) and slowly vortexing the mixture until all of the CA was dissolved, Filter into a new eppendorf tube or centrifuge at 4000 rpm for 5 minutes and transfer the supernatant to a new eppendorf tube to remove any aggregate / precipitate material. The amount of 10 mg / ml Tween 20 stock solution required to bring the final concentration of Tween 20 to 0.5 mg / ml in the final reaction mixture was added. Add the required amount of G-CSF protein solution to the CA solution to obtain an 11 molar excess of CA (relative to 40 kDa) and mix slowly by placing the reaction mixture on a gentle shaker at 4 ± 1 ° C. I let you. To reach 50 mM or 3.17 mg / ml in the final reaction mixture, add 100 mg / ml NaCNBH ₃ solution and mix slowly to check the pH of the final reaction mixture, if necessary 4 ± 1 ° C. The pH was adjusted to 5.5 with 1 M NaOH / HCl. Finally, 10 mM NaOAc, 5% sorbitol (pH 5.5) was used to adjust the reaction volume to obtain a protein concentration of 0.67 mg / ml in the reaction mixture. The tube was sealed and stirred at the desired temperature (4 ± 1 ° C.) for 24 hours. Stop the reaction in an appropriate manner, collect the sample for in vitro activity analysis in MNFS 60 cell, SDS-PAGE (using 4% -20% trisglycine gel), SE-HPLC; superose 6 column) and react The pH of the mixture was checked. To remove any precipitate, the reaction mixture was centrifuged at 13000 rpm for 5 minutes before SE-HPLC analysis and purification. The preferred buffer for SE-HPLC was 0.1 M Na phosphate (pH 6.9). 3.2 Purification and Characterization of GCSF-CA Complex The remaining reaction mixture sample was diluted with AEX buffer A (20 mM sodium acetate, 50 mM sodium chloride, pH 5.0) (1.5 ml reaction mixture + buffer A 9 ml), the pH was checked, adjusted to 5.0 if necessary, and loaded onto an AEX column pre-equilibrated with AEX buffer A. The packed fraction was collected and labeled. The column was washed with AEX buffer A (at least 5 column volumes) and fractions (1.5 column volumes for each fraction) were collected and labeled. The product was eluted with AEX buffer B (50 mM sodium phosphate, 0.65 M sodium chloride, pH 7.0) and fractions (1 column volume fraction, 6 column volumes each) were collected and labeled. . If two consecutive fractions were not present in the protein content (UV280 nm), the next step was taken. Samples were placed on ice during purification. Protein concentration was analyzed by UV (280 nm) (Abs of 1 mg / ml G-CSF was about 0.872). Samples were collected for SDS-PAGE and SE-HPLC. HIC was used to remove free CA from the mixture. Samples were concentrated as needed.

AEX fraction-containing complexes were pooled and (NH ₄ ) ₂ SO ₄ was added to a concentration of 2.75 M in the loading solution. The solution was then loaded onto a HIC column pre-equilibrated with HIC buffer A (10 mM sodium phosphate, 2.75 M ammonium sulfate (pH 6.5)). The packed fractions (1.5 column volume fractions each) were collected and labeled. The column was washed with HIC buffer A (at least 5 column volumes, flow rate = 0.5 ml / min) and fractions (1.5 column volumes) were collected and labeled. The product was eluted with HIC buffer B (20 mM sodium phosphate pH 7.4) (flow rate = 5 ml / min) and fractions (1 column volume fraction, 6 column volumes) were collected and labeled. Samples were placed on ice during purification. Protein concentration was analyzed by UV (280 nm). Combining the HIC fractions containing the purified complex, the complex composition in solution was adjusted with 50% sorbitol solution and 10 mg / ml Tween 20 solution, and the final composition of 5% sorbitol and 0.025 mg / ml Tween 20 I got a thing. The solution was then concentrated at 4 ± 1 ° C. and analyzed for protein concentration by UV (280 nm). Further purification by SE-HPLC is also possible (eg to separate the complex from free protein / aggregates etc.). The complex was sterile filtered and samples were collected for activity analysis and characterization by SDS-PAGE and SE-HPLC. Aliquots were removed for protein analysis and CA analysis as needed. The remainder was stored at 4 ± 1 ° C. until next use and tested for physical stability by SE-HPLC.

The effects of various processes affecting the stability and degree of derivatization of GCSF in solution were tested.
Results Granulocytic-colony stimulating factor (G-CSF) colominic acid (CA) complex selectively (at 20 mg scale) at the N-terminus by performing the reaction at low pH (pH 5.5) and 4 ± 1 ° C. Techniques for preparing and purifying are detailed above. This involves binding in the presence of sodium cyanoborohydride, followed by purification using ion exchange chromatography (AEX) to remove free G-CSF, followed by removal of CA by hydrophobic interaction chromatography (HIC). Accompany. Low pH was used to favor selective derivatization of the N-terminal α-amino group and also to minimize GCSF aggregation during the reaction. The composition of the final reaction buffer was 5% sorbitol, 0.5 mg / ml Tween 20 in 10 mM NaOAc (pH 5.5).

  SE-HPLC (change in retention time of GCSF-CS compared to GCSF, both parts also elute simultaneously), ion exchange chromatography (binding of complex on AEC column) and polyacrylamide gel electrophoresis (SDS-PAGE) The formation of a GCSF-CA complex was confirmed by the shift of the band of a high molecular weight species). The complex used in the in vitro cell line analysis (in MNFS-60 cells) was about 40% active compared to the native protein. Proteins with low derivatization aggregated with the complex prepared without formulation additives. The left side of FIG. 3.1 shows SE-HPLC data after 24 hours for a 39 kDa reaction mixture of GCSF-CA prepared in the presence of Tween20. The characterization conditions were Superdex 200, 0.15 M ammonium bicarbonate (pH 7.8). Table 2 shows the peak analysis.

The right side of Fig. 3.1 shows the result of SDS-PAGE.
4.1 Preparation of N-terminal leptin-CA complex with formulation additives Leptin was given as a lyophilized powder (molecular weight 16240) and stored at -80 ° C. The amount of colominic acid (eg oxidized or non-oxidized colominic acid, 7.5 molar excess) added for binding was calculated. Colominic acid was dissolved in a minimum amount of sodium acetate buffer and filtered to adjust the pH to 5.5. After adding the colominic acid solution in leptin solution (20 mM sodium acetate, 1% sucrose, 10 mM L-glutamic acid and protein in 0.01% Tween 20 (pH 5.5)), 50 mM or 3.17 mg in the reaction mixture. The necessary μl of NaCNBH ₃ required to make / ml was added, mixed slowly, the pH of the final reaction mixture was checked and the pH was adjusted to 5.5. The tube was sealed and stirred at the desired temperature (4 ± 1 ° C.) for 24 hours. After incubation, necessary samples were collected (for SDS-PAGE, SE-HPLC, etc.). Protein concentration was analyzed by UV (280 nm) (absorbance of lmg / ml leptin is 0.878). Various processes with different effects on protein derivatization and stability were tested.
4.2 Purification and Characterization of Leptin-CA Complex HIC and IEC were used to remove excess CA and free leptin from the reaction mixture, respectively. Aliquots were removed for protein analysis and CA analysis. The remainder was stored at −80 ° C. until use. The product was characterized by SDS-PAGE, SE-HPLC, Western blotting, CA and protein analysis and the like.
Results The procedure for preparing and purifying the colominic acid (CA) complex of leptin selectively (at the 5 mg scale) at the N-terminus by reacting at low pH (pH 5.5) and 4 ± 1 ° C. is shown above. Has been. This involves binding in the presence of sodium cyanoborohydride, followed by purification using ion exchange chromatography (AEX) to remove free G-CSF, followed by removal of CA by hydrophobic interaction chromatography (HIC). Accompany. Low pH and formulation additives were used in favor of selective derivatization of the N-terminal α-amino group and also to minimize aggregation of GCSF during the reaction. The composition of the final reaction buffer was protein in 20 mM sodium acetate, 1% sucrose, 10 mM L-glutamic acid and 0.01% Tween 20 (pH 5.5).

  SE-HPLC (change in retention time of leptin-CA compared to leptin, both parts also elute simultaneously), ion exchange chromatography (binding of complex to AEC column) and polyacrylamide gel electrophoresis (SDS-PAGE) The formation of leptin-CA complex was confirmed by the shift of the high molecular weight species band). Proteins with low derivatization aggregated with the complex prepared without formulation additives.

The left side of FIG. 4.1 shows SDS-PAGE of the 50 kDa CAO-leptin complex, and the right side shows the results of SEC-HPLC.
5.1 Preparation of N-terminal erythropoietin (EPO) -CA complex EPO is given as a solution (0.34 mg / ml in 10 mM phosphate buffer, 130 mM NaCl, pH 7.0, specific activity: 110,000 U / ml, molecular weight 30600) and stored at −32 ° C. The protein was thawed at 2 ° C. to 8 ° C. and the required amount was placed in a 2 ml Eppendorf tube. The required amount of colominic acid was taken and the protein solution was added to solid CA and mixed slowly. Add μl of sodium cyanoborohydride solution required to reach 50 mM or 3.17 mg / ml in the reaction mixture, vortex to check the pH of the final reaction mixture, and adjust the pH to 7. 4 was adjusted. The tube was sealed and stirred at the desired temperature (4 ± 1 ° C.) for 24 hours. After incubation, the necessary samples were collected (eg for activity analysis, SDS-PAGE, SE-HPLC).
5.2 Purification and characterization of the epo-CA complex The remaining reaction mixture sample was diluted with HIC buffer A (1.2 M ammonium sulfate, pH 6.3) (1 ml sample + 4 ml buffer A) and then with HIC buffer A. A pre-equilibrated HIC column was packed. The packed fraction was collected and labeled. The column was washed with HIC buffer A (1.2 M ammonium sulfate (pH 6.3)) (at least 10 ml) and fractions were collected and labeled. The product was eluted with HIC buffer B (10 mM Tris buffer (pH 7.0)) and the first fraction (0.5 ml) was collected and the subsequent 0.5 ml to 1 ml fractions were labeled. Samples were kept on ice during purification.

  Protein concentration was analyzed by UV (280 nm) (Abs of 1 mg / ml EPO was about 0.743). Samples were taken for SDS-PAGE. When the molecular weight of CA was too small (for example, 22 kDa) to separate the complex from EPO by SE-HPLC, unbound EPO was separated using anion exchange chromatography (AXC). For AXC, the HIC fraction-containing protein was diluted with AXC buffer A (10 mM Tris buffer (pH 7.0)) (sample 1 ml + AXC buffer A 5 ml) and loaded onto an AXC column pre-equilibrated with AXC buffer A. . The packed fraction was collected and labeled. The column was washed with AXC buffer B (20 mM sodium acetate (pH 3.0)) (at least 10 ml) and fractions were collected and labeled. Elute the product with AXC buffer C (50 mM sodium phosphate buffer, 1 M sodium chloride, pH 7.0) and collect the first fraction (0.5 ml) followed by 0.5 ml to 1 ml fractions. Labeled. Samples were kept on ice during purification.

  Alternative purification can be performed by SE-HPLC (for example to separate the complex from EPO if the CA used has a high molecular weight (eg 39 kDa)). Protein concentration was analyzed by UV (280 nm) (Abs of 1 mg / ml EPO is about 0.743). Samples were collected for SDS-PAGE.

Aliquots were removed for protein analysis and CA analysis. The remainder was stored at -20 ° C until use. The product was characterized by SDS-PAGE. To measure the activity of EPO, samples in erythroid progenitors that induce proliferation in vitro, isolated from the spleen of mice artificially anemic by injection of phenylhydrazine, were used. The protocol was adjusted based on the method reported in Krystal (1972). This analysis is by measuring the rate of DNA replication by adding EPO to erythroid progenitor cells and determining the rate of ³ H-thymidine incorporation. In vivo pharmacokinetic (PK) and pharmacodynamic (PD) studies were performed on B6D2F1 mice.
Results SE-HPLC (change in retention time of EPO-CA compared to EPO, both parts also elute simultaneously), ion exchange chromatography (complex binding to AEC column) and polyacrylamide gel electrophoresis (SDS- The formation of the EPO-CA complex was confirmed by PAGE (shift of the high molecular weight species band) (FIG. 5.3). The polysialized sample was active in vitro and showed a very good profile (PK and PD) relative to plain EPO. Figures 5.1 and 5.2 show the in vivo results.

  The left side of Figure 5.3 shows the SE-HPLC of the 39 kDa complex of EPO-CA after 24 hours. Table 3 is a peak analysis table. Characterization conditions: column Superdex 200, buffer 0.15 M ammonium bicarbonate (pH 7.8).

6.1 Preparation of N-terminal non-glycosylated (NGepo-CA) complex Naked EPO is given as a solution (0.18 mg / ml in 20 mM sodium phosphate buffer, 300 mM NaCl, pH 6.65). ml, specific activity: 100,000 U / ml, molecular weight 19000), stored at −32 ° C. The protein was thawed at 2 ° C. to 8 ° C. and the required amount was placed in a 2 ml Eppendorf tube. The amount of colominic acid (eg oxidized or non-oxidized colominic acid) added for binding was calculated. The required amount of colominic acid was weighed and the weight was recorded. The protein solution was added to the solid CA and mixed slowly. Add μl of sodium cyanoborohydride solution required to reach 50 mM or 3.17 mg / ml in the reaction mixture, vortex to check the pH of the final reaction mixture, and adjust the pH to 7. 4 was adjusted. The tube was sealed and stirred at the desired temperature (4 ± 1 ° C.) for 24 hours. After incubation, the necessary samples were collected (eg for activity analysis, SDS-PAGE, SE-HPLC).
6.2 Purification and Characterization of NGepo-CA Complex The remaining reaction mixture sample was diluted with HIC buffer A (1.2 M ammonium sulfate (pH 6.3)) (1 ml sample + 4 ml buffer A) and then with HIC buffer A. A pre-equilibrated HIC column was packed. The packed fraction was collected and labeled. The column was washed with HIC buffer A (at least 10 ml) and fractions were collected and labeled. The product was eluted with HIC buffer B and the first fraction (0.5 ml) was collected and the subsequent 0.5 ml to 1 ml fractions were labeled. Samples were kept on ice during purification. Protein concentration was analyzed by UV (280 nm) (absorbance of 1 mg / ml nEPO was about 0.743). Samples were collected for SDS-PAGE. Depending on the reaction conditions, there was no need for further purification since there was no untreated free EPO in the reaction mixture. If untreated EPO was present in the reaction mixture, Vivaspin 6 (5000 MWCO) can be used to concentrate the HIC fraction-containing protein and purification can be performed by SE-HPLC. Protein concentration was analyzed by UV (280 nm) (absorbance of 1 mg / ml nEPO is about 0.743). Samples were collected for SDS-PAGE.

Aliquots were removed for protein analysis and CA analysis. The remainder was stored at -20 ° C until use. The product was characterized by SDS-PAGE.
Results SE-HPLC (change in retention time of NGpo-CA compared to NGEPO, both parts elute simultaneously), ion exchange chromatography (binding of complex on AEC column) and polyacrylamide gel electrophoresis (SDS- The formation of the NGepo-CA complex was confirmed by PAGE (shift of a high molecular weight species band) (FIGS. 6.1 to 6.3). The left side of FIG. 6.3 shows the 39 kDa complex of EPO-CA after 24 hours. The polysialized sample was active in vitro and showed a very good profile (PK and PD) relative to untreated NGepo.

  FIG. 6.3 shows the results of SE-HPLC. Peak analysis is shown in Table 4 below. Characterization conditions-column Superdex 200, buffer: 0.15 M ammonium bicarbonate (pH 7.8).

Fig. 6.1 shows the results of in vivo clearance. PSA-NGEPO showed a very good profile compared to NGEPO.
7.1 Preparation of N-terminal insulin-CA complex Insulin was dissolved in a minimum of 100 mM HCl and then adjusted to the required pH. The amount of colominic acid (eg, oxidized or non-oxidized colominic acid) for binding was calculated. The required amount of colominic acid was weighed, dissolved in the minimum amount of reaction buffer, added to the protein solution and mixed gently using a vortex mixer. Μl of sodium cyanoborohydride required to obtain a final concentration of 4 mg per ml of reaction mixture was added. Suitable stabilizers were used in the reaction mixture as needed. The tube was sealed and stirred for 48 hours at the desired temperature (37 ° C. as required). The time and temperature may be varied depending on the protein used. Polysialylated protein was purified by IEC and HIC. After 24 hours, the protein-polymer complex reached 100%. This was characterized by native-page, SDS-page, size exclusion chromatography, ion exchange chromatography and the like.

In IEC, the insulin and insulin-colominic acid complex is purified by cation exchange chromatography on a High Trap SP column to separate polysialylated insulin (which does not bind to the column) from free insulin (which remains on the column). This involves separation of the complex below the isoelectric point of 5.2 (at least one unit below) using a cation exchange resin. The activity of these complexes was measured in healthy mice. 7.2 Purification and Characterization of Insulin-CA Complex A sample of the reaction mixture is diluted 5-fold with AEX buffer A (0.05 M sodium acetate (pH 4.4)), the pH is checked, and the pH is adjusted as necessary. Was loaded to 4.4 and loaded onto an AEX column (flow rate = 1 ml / min) pre-equilibrated with AEX buffer A. The packed fractions (1.5 column volume fractions each) were collected and labeled. The column was washed with AEX buffer A (0.05 M sodium acetate (pH 4.4)) (at least 5 column volumes, flow rate = 1 ml / min), and fractions (1.5 column volume fractions each) were collected. Labeled. The product was eluted with AEX buffer B (0.05 M sodium acetate, 1 M sodium chloride, pH 4.4) (flow rate = 1 ml / min) and fractions (1 column volume fraction, 6 columns each). Collected and labeled. If two consecutive fractions were not present in the protein content, proceeded to the next step. Samples were kept on ice during purification.

AEX fraction-containing complexes are pooled, diluted 10-fold with HIC buffer A (0.8 M ammonium sulfate, 50 mM sodium phosphate (pH 7.4)), and pH is adjusted to 7. with hydrochloric acid solution or sodium hydroxide solution. 4 was adjusted. The solution was then loaded onto a HIC column (flow rate = 0.3 ml / min) pre-equilibrated with HIC buffer A (0.8 M ammonium sulfate, 50 mM sodium phosphate, pH 7.4). The packed fractions (1.5 column volume fractions each) were collected and labeled. The column was washed with HIC buffer A (at least 5 column volumes, flow rate = 0.5 ml / min) and fractions (1.5 column volumes) were collected and labeled. The product was eluted with HIC buffer B (50 mM sodium phosphate pH 7.4) (flow rate = 5 ml / min) and fractions (1 column volume fraction, 6 column volumes) were collected and labeled. Samples were kept on ice during purification. Protein concentration was analyzed by UV (280 nm). The HIC fractions containing the purified complex were combined and the solution was concentrated at 4 ± 1 ° C. and analyzed for protein concentration by UV (280 nm).
Results SE-HPLC (change in retention time of insulin-CA compared to insulin, both parts also elute simultaneously), ion exchange chromatography (elution of complex on CEC column) and polyacrylamide gel electrophoresis (SDS- The formation of insulin-CA complex was confirmed by PAGE (shift of high molecular weight species band) (FIGS. 7.1 and 7.2). The polysialylated sample showed superior in vivo efficacy compared to the native protein.

In FIG. 7.2, the HPLC conditions are as follows: column Superose 12, buffer: 0.1 M sodium phosphate (pH 6.9), flow rate 0.25 ml / min, injection volume 200 μl.
8.1 Preparation of N-terminal interferon-CA complex The procedure for preparing and purifying the colominic acid (CA) complex of IFNα2b is the HIC to remove the subsequent free colominic acid in the presence of sodium cyanoborohydride. Followed by removal of unbound IFN by either AXC or SE-HPLC (optional) (eg 1 mg scale).

  IFNα2b was given as a solution (1.75 mg / ml in acetate buffer (pH 5)) and stored at −32 ° C. The protein was thawed at 2 ° C. to 8 ° C. and the required amount was placed in a 2 ml Eppendorf tube. If the protein concentration in the reaction mixture was lower than 1.75 mg / ml, it was diluted with the required amount of PBS (pH 7.4).

The required amount of CA was weighed and recorded. CA was solubilized in the minimum amount of reaction buffer, added to the protein solution and mixed gently using a vortex. The μl required to reach 50 mM or 3.17 mg / ml in the reaction mixture was added and mixed gently to check the pH of the final reaction mixture and adjust the pH to 6.0 if necessary. The tube was sealed and stirred at the desired temperature (4 ± 1 ° C.) for 24 hours. After the incubation period, necessary samples were collected (eg for activity analysis, SDS-PAGE, SE-HPLC).
8.2 Purification and Characterization of Interferon-CA Complex The remaining reaction mixture sample was diluted with HIC buffer A (25 mM Tris buffer, 3 M sodium chloride (pH 7.5)) (sample 1 ml + buffer A 4 ml), An HIC column pre-equilibrated with HIC buffer A was packed. The packed fraction was collected and labeled. The column was washed with HIC buffer A (25 mM Tris buffer, 3 M sodium chloride (pH 7.5)) (at least 10 ml) and fractions were collected and labeled. The column was eluted with HIC buffer B (25 mM Tris buffer (pH 7.5)), the first fraction (0.5 ml) was collected, and the subsequent 0.5 ml to 1 ml fractions were collected and labeled. Samples were kept on ice during purification.

  Protein concentration was analyzed by UV (280 nm) (absorbance of 1 mg / ml IFN is about 1). Samples were collected for SDS-PAGE.

  Separation of unbound IFN was performed using anion exchange chromatography (AXC) or SE-HPLC. For AXC, the HIC fraction-containing protein was diluted with AXC buffer A (25 mM Tris buffer (pH 7.5)) (sample 1 ml + AXC buffer A 5 ml) and loaded onto an AXC column pre-equilibrated with AXC buffer A. . The packed fraction was collected and labeled. The column was washed with AXC buffer B (50 mM sodium phosphate, 150 mM sodium chloride (pH 5)) (at least 10 ml) and fractions were collected and labeled. The product was eluted with AXC buffer C (50 mM sodium phosphate, 1 M sodium chloride, pH 7) and the first fraction (0.5 ml) was collected and the subsequent 0.5 ml to 1 ml fractions were labeled. . Samples were kept on ice during purification.

  Further purification was performed by SE-HPLC (eg to separate the di-complex from the mono-complex) (FIG. 8.1).

Protein concentration was analyzed by UV (280 nm) (absorbance of 1 mg / ml IFN is about 1) or BCA analysis. Samples were collected for SDS-PAGE. Aliquots were removed for protein analysis and CA analysis. The remaining product was stored at −20 ° C. until use. The product was characterized by SDS-PAGE. Activity was determined with a Daudi cell line.
Results SE-HPLC (change in retention time of interferon-CA compared to interferon, both parts also elute simultaneously), ion exchange chromatography (binding complex to AEC column) and polyacrylamide gel electrophoresis (SDS- The formation of an interferon-CA complex was confirmed by PAGE (shift of a high molecular weight species band) (FIGS. 8.1 to 8.2). The polysialylated sample was active in vitro and showed a very good profile (PK) relative to untreated interferon.

  Table 5 shows the peak analysis of SE-HPLC in FIG.

9.1. Polyesterification of obestatin Obestatin is a 2.5 kDa appetite suppressor hormone. Although encoded by the ghrelin gene, obestatin antagonizes the appetite stimulating effect of ghrelin. Obestatin treatment of rats has been shown to reduce food intake, inhibit jejunal contraction, and reduce weight gain.
9.2 N-terminal binding (site specific)
A 15 molar excess of 14 kDa oxidized polysialic acid (PSA) was dissolved in the buffer and the pH was adjusted to 6.0. Then obestatin and 50 mM (final concentration) sodium cyanoborohydride were added and the pH was adjusted again to bring the reaction mixture to the required volume. The reaction was performed at 4 ± 1 ° C. with gentle shaking for 18 hours.
9.3 Random binding (relative)
A 10 molar excess of 14 kDa oxidized PSA was dissolved in the buffer and the pH was adjusted to 7.4. Then obestatin and 50 mM (final concentration) sodium cyanoborohydride were added and the pH was readjusted to bring the reaction mixture to the required volume. The reaction was performed at 4 ± 1 ° C. with gentle shaking for 18 hours.
Analysis of binding Confirm binding by size exclusion high performance liquid chromatography (SE-HPLC) by reducing retention time and thus increasing size. At a flow rate of 0.25 mL / min, 100 μL of the binding reaction was injected onto an SE-HPLC Superose 12 column pre-equilibrated with 0.1 M sodium phosphate (pH 6.9). The absorbance at 280 nm was recorded.

  Complexes are visualized by reduced mobility on sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) gels (FIG. 9). Samples were electrophoresed on 4% -20% trisglycine gels under non-denaturing conditions. Lane 1: molecular weight marker, 2: obestatin, 3: reaction at pH 6.0, 4: reaction at pH 7.4.

Obestatin-PSA conjugates were successfully made at both pH 6.0 and pH 7.4, as confirmed by SE-HPLC. The complex made at pH 7.4 was also visualized by SDS-PAGE.
10.1 DNase I Polysialyation In patients with cystic fibrosis (CF), the retention of viscous purulent secretions in the airways contributes to reduced lung function and exacerbated infection. The pulmonary purulent secretion contains a very high concentration of extracellular DNA that is released by degenerating leukocytes that accumulate in response to infection. Treatment of CF patients with deoxyribonuclease I (DNase I) hydrolyzes the DNA, thereby reducing vaginal viscoelasticity. This treatment has also been proposed for the treatment of systemic lupus erythematosus and tumor targeting.
10.2 N-terminal binding (site specific)
A 20 molar excess of 26 kDa oxidized polysialic acid (PSA) was dissolved in the buffer and the pH was adjusted to 6.0. Then bovine DNase I (Samsong & Sigma) and 50
mM (final concentration) sodium cyanoborohydride was added and the pH was adjusted again to bring the reaction mixture to the required volume. The reaction was performed at 37 ± 1 ° C. and 4 ± 1 ° C. with gentle shaking for 18 hours.
10.3 Random binding (relative)
A 10-50 molar excess of 14 kDa oxidized PSA was dissolved in the buffer and the pH was adjusted to 7.4. Then bovine DNase I (Samsong & Sigma) and 50 mM (final concentration) sodium cyanoborohydride were added and the pH was adjusted again to bring the reaction mixture to the required volume. The reaction was performed at 37 ± 1 ° C. and 4 ± 1 ° C. with gentle shaking for 18 hours.
10.4 Binding analysis Complexes were visualized by reduced mobility on sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) gels (Figure 10.1). Samples were electrophoresed on 4% -20% trisglycine gels under non-denaturing conditions. Lane 1: molecular weight marker, 2: blank, 3: DNase I, 4: Sigma DNase I, 5: blank, 6: DNase I with non-oxidized PSA, 7: DNase I oxidized PSA (pH 7.4) ) Reactant, 8: DNase I oxidized PSA (pH 6.0) reactant.
10.5 Complex Purification DNase I-PSA complex using hydrophobic interaction chromatography (HIC) -phenyl sepharose matrix, starting buffer containing 2.0 M ammonium sulfate, elution in buffer without ammonium sulfate Was purified. The eluted fraction was then added to the ion exchange matrix Q-Sepharose Fast Flow and eluted with a buffer containing sodium chloride. Purification of the complex was confirmed by size exclusion high performance liquid chromatography (SE-HPLC) (FIG. 10.2). 100 μl of the binding reaction was injected onto a SE-HPLC Superose 12 column pre-equilibrated with 0.1 M sodium phosphate (pH 6.9) at a flow rate of 0.25 mL / min. The absorbance at 280 nm was recorded.

The DNase I-PSA complex shows a reduction in retention time in the size exclusion column compared to DNase I alone.
10.6 Purification of active DNase I Active DNase I was purified from a mixture of active DNase I and heat-inactivated DNase I by heparin-sepahrose chromatography. The DNase I mixture was added to the column in low salt buffer and eluted with an increasing gradient of sodium chloride. The activity per mg DNase I in fraction A12 was approximately 4 times higher than that of the sample.
10.7 Activity of the complex The activity of the complex was measured and unbound DNase I using methyl-green analysis (Sinicropi et. Al., (1994) Anal Biochem, 222 (2): 351-8). Compared to the activity of 37 ± 1 ℃
The purified DNase I-PSA complex produced in Example ₃ showed approximately 10% activity in 4 mM CaCl ₂ buffer (see below) compared to unbound DNase I.

The effect of CaCl ₂ titration on the DNA hydrolysis activity of DNase I and DNase I-PSA complex was measured (n = 1) (FIG. 10.3). CaCl ₂ is a formulation additive. A methyl-green DNase I analysis was set up and more CaCl ₂ was added. Activity is expressed relative to activity in 4 mM CaCl ₂ buffer. A methyl-green analysis of DNase I was set up with bound DNase I and free DNase I, and CaCl ₂ was added at various concentrations up to a concentration of 100 mM. It was found that the addition of CaCl ₂ increased the activity of bound DNase I over that of free DNase I. Thus, there is a need for more CaCl ₂ for full activation of DN DNase -PSA complex.
Conclusion An active DNase I-PSA complex has been made in the presence of sodium cyanoborohydride. The complex is purified from free DNase I and shows an activity of approximately 10% per mg of DNase I compared to the activity of free DNase I.
References

Claims (29)

  A method for producing a purified N-terminal derivative of a protein or peptide, comprising: (i) reacting a polysaccharide with an amine-terminal amine group of the protein or peptide in an acidic aqueous solution to produce an N-terminal derivative; (ii) A method for producing a purified N-terminal derivative of a protein or peptide, wherein the obtained N-terminal derivative is purified in an aqueous solution having a higher pH than in step (i).   The method of claim 1, wherein the polysaccharide has a reactive aldehyde group and step (i) is carried out under reducing conditions.   The method of claim 2, wherein the reactive aldehyde group is at a non-reducing end of the polysaccharide.   The method according to any one of claims 1 to 3, wherein the polysaccharide comprises at least one sialic acid unit or a portion derived from a sialic acid unit.   The method according to claim 4, wherein the polysaccharide comprises a sialic acid unit or a portion derived from a sialic acid unit at its non-reducing end and / or reducing end.   The method according to claim 4 or 5, wherein the polysaccharide is a polysialic acid consisting essentially of sialic acid units.   7. A method according to claim 6, wherein the polysaccharide has at least 2, preferably at least 5, more preferably at least 10, most preferably at least 50 sialic acid units in the molecule.   8. The reactive aldehyde is at the reducing end of the polysaccharide and the non-reducing end is passivated such that it does not react with the N-terminus of the protein. The method described in 1.   The pH of the aqueous solution in step (i) is in the range of 4.0 to 6.0, and the pH of the aqueous solution in step (ii) is in the range of 6.5 to 8.5. The method according to any one of the above.   The method according to any one of claims 1 to 9, wherein the protein is a therapeutic protein.   The method according to any one of claims 1 to 10, wherein the protein is leptin, interferon, FSH, galactosidase, or DNase.   The method according to any one of claims 1 to 11, wherein a formulation additive is added to the acidic aqueous solution in step (i).   13. The method of claim 12, wherein the formulation additive is selected from one or more of buffers, stabilizers, surfactants, salts, polymers, metal ions, sugars, polyols or amino acids.   14. The method of claim 13, wherein the formulation additive is sorbitol, trehalose or sucrose.   14. The method of claim 13, wherein the formulation additive is a nonionic surfactant.   14. A method according to claim 13, wherein the formulation additive is a polymer selected from PSA, PEG or hydroxy- [beta] -cyclodextrin. 14. A method according to claim 13, wherein the formulation additive is a divalent metal ion, preferably Zn2 ⁺ , Ni2 ⁺ , Co2 ⁺ , Cu2 ⁺ , Sr2 ⁺ , Fe2 ⁺ , Ca2 ⁺ or Mg2 ⁺ .   14. The method of claim 13, wherein the formulation additive is a buffer and the buffer is sodium phosphate / sodium acetate.   The method according to any one of claims 2 and 3 to 18, wherein the reduction condition is obtained by the presence of borohydride in the acidic aqueous solution in step (i).   A composition comprising a polysialic acid derivative population of a protein, wherein the derivative comprises 2 to 200 sialic acid units, the population consisting essentially of only the N-terminal derivative of the protein A composition comprising a population.   21. The composition of claim 20, wherein the protein is leptin and the derivative comprises 100 to 150 sialic acid units.   21. The composition of claim 20, wherein the protein is FSH and the derivative comprises 75 to 200 sialic acid units.   21. The composition of claim 20, wherein the protein is [alpha] -galactosidase and the derivative comprises 20 to 150 sialic acid units.   21. The composition of claim 20, wherein the protein is a DNase and the derivative comprises 2 to 120 sialic acid units.   21. The composition of claim 20, wherein the protein is IFN and the derivative comprises 80 to 180 sialic acid units.   21. The composition of claim 20, wherein the protein is obestatin and the derivative comprises 20-100 sialic acid units.   26. The composition according to any one of claims 20 to 25, wherein the polysialic acid has a polydispersity of less than 1.3, preferably less than 1.2, and most preferably less than 1.1.   28. A composition according to any one of claims 20 to 27, which is a pharmaceutical composition and further comprises one or more pharmaceutically acceptable excipients.   An N-terminal derivative of a protein or peptide obtainable by the method according to any one of claims 1-19.

Images